Author(s): SAMPIERI, A., SPINELLO, G., GOLDIN, E., MASARIN, A., FRANCHI, M.V., PAOLI, A., MORO, T., CASOLO, A., Institution: UNIVERSITY OF PADUA, ITALY, Country: ITALY, Abstract-ID: 2166

INTRODUCTION:
High-density surface EMG (HDsEMG) offers valuable non-invasive insights into individual motor unit (MU) activity and properties. However, large intersubject variability in the number of identified MU is often observed due to specific neural aspects and anatomical factors (e.g. volume conductor properties)1,2. A recent study3 on young males revealed that greater muscle electrode distance (MED), which includes skin, subcutaneous fat and superficial muscle aponeurosis, adversely affects HDsEMG signal decomposition in biceps brachii, especially at low force levels. Here, we investigated the influence of body composition and anatomical features on the number of identified MU in the human vastus lateralis (VL) muscle.
METHODS:
To date, 48 healthy participants (17% female) representing two age categories (young adults (YG): 19-30 yr., elderly: 66-82 yr.) were enrolled in this study. They performed submaximal isometric knee extensions at 15%, 35%, 50%, and 70% of maximal voluntary contraction (MVC), while HDsEMG recorded the activity of the VL of the right leg (RL). HDsEMG signals were decomposed into individual MU for each force level. Multiple body composition features were evaluated using BIA and DXA, while ultrasonography (US) was used to measure MED precisely below the HDsEMG electrodes. Correlations and regression analyses were used to assess the relationships between the number of detected MU and body composition features according to force levels.
RESULTS:
A total of 1476 unique MU were detected from the VL. Significantly negative correlations (Spearman’s, p<.05 in all cases) were observed between MU number and fat mass (FM) estimated by BIA (15%: r=-.33; 50%: r=-.32; 70%: r=-.35) and DXA (15%: r=-.36; 35%: r=-.36; 50%: r=-.32; 70%: r=-.39), as well as FM RL estimated by DXA (15%: r=-.50; 35%: r=-.58; 50%: r=-.59; 70%: r=-.59), and MED (15%: r=-.69; 35%: r=-.72; 50%: r=-.72; 70%: r=-.72). Conversely, no associations were found between any other anthropometrical variable and the number of detected MU. When stratified by age group, simple linear regression analysis revealed that MED alone explained 79%, 66%, 61%, and 64% of the variance in the number of detected MU at 15%, 35%, 50%, and 70% MVC, respectively, in elderly, whereas 38%, 37%, 30%, and 32% at the same force levels in YG.
CONCLUSION:
Our findings further confirm the influence of FM, and particularly MED (mainly composed of subcutaneous fat), on the quality of HDsEMG signal decomposition among healthy individuals with diverse characteristics. Specifically, greater body FM assessed by BIA and DXA were associated with decreased detectable MU. Notably, a more localized analysis of the anatomical area underlying HDsEMG electrodes enhanced the accuracy of MU number prediction. Our results highlight the importance of assessing MED using the US prior to HDsEMG recordings, as it emerges as the primary predictive parameter for MU number identification.
1Del Vecchio et al 2020
2Farina and Holobar 2016
3Souza de Oliveira et al 2022